Preterm birth remains the most pressing public health problem in obstetrics, affecting 11.54% of U.S. newborns in 2012, with direct costs of at least $65,600 per neonate.1,2 Recently interventions such as progesterone and cervical cerclage have been shown to decrease the rate of preterm birth among at-risk women.3–5 Nonetheless, the effect of these efforts has been limited by a lack of accurate predictors of preterm birth.6–8
It has long been known that the fetal adrenal gland plays an integral role in both term and preterm parturition.9 For example, necropsy studies of preterm fetuses demonstrated that the gross weight of the fetal adrenal gland is significantly greater among those fetuses that undergo spontaneous preterm birth compared with those delivered for maternal reasons.10 Likewise, it has been noted that the “fetal zone” of the adrenal gland is greatest in size immediately before spontaneous birth and quickly involutes after delivery.11
These findings led a group of investigators to examine the role of ultrasound measurement of the fetal adrenal gland in the acute prediction of spontaneous preterm birth. In a cohort of 126 women with and without symptoms of preterm labor, Turan et al12 measured the volume of the fetal adrenal gland to determine whether it differentiated women who went on to deliver in the next 5 days. They found that women who delivered in the next 5 days had larger adrenal volumes compared with those who did not. In a subsequent publication, Turan et al13 examined the fetal adrenal gland in 74 women with preterm labor symptoms using a simplified ultrasound technique examining the ratio of the depth of the fetal adrenal zone (depth) to the depth of the total gland (Depth). Similar to their prior experience, they found that this measurement discriminated women who went on to deliver within 7 days and those who did not.
Recognizing that the process of parturition occurs over weeks to months before delivery14 and that this extended timeline may provide an opportunity to clinically intervene, we sought to examine whether ultrasound measurement of the fetal adrenal gland can accurately identify asymptomatic nulliparous women who will undergo spontaneous preterm birth. Our primary hypothesis was that the measurement of the ratio of the fetal zone to the overall adrenal gland will discriminate fetuses destined to deliver prematurely from those who will deliver at term.
MATERIALS AND METHODS
The Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be Network was formed with the goal of predicting adverse pregnancy outcomes, including spontaneous preterm birth, among nulliparous women. Eight institutions recruited 10,038 nulliparous women with singleton pregnancies into a prospective observational study that involved serial evaluation at three separate time periods during pregnancy. Before the initiation of the study, institutional review board approval was obtained from each clinical site and the data coordinating center. Further description of the methods of the study can be found in the publication by Haas et al.15
Before the initiation of the adrenal ultrasound portion of the study, study ultrasonographers were required to demonstrate competency in both identifying and measuring the adrenal gland in a transverse plane and, when possible, in the coronal or in the sagittal plane. These skills were taught through mandatory attendance of a webinar and, where required, hands-on training with a credentialed ultrasonographer. As a result of technical difficulty of consistently obtaining high-quality sagittal adrenal gland measurements as described in the prior publication by Turan et al,13 a decision was made to simplify the technique and primarily examine the fetal adrenal gland in the transverse plane. Ultrasonographers were encouraged, but not required, to become credentialed in measuring the depth of the gland in either the coronal or sagittal plane as well.
A modification of the previously described technique of measuring the ratio of the fetal zone/total gland measurement in the sagittal plane, measurement of the fetal zone/total gland measurement in the transverse plane (length and width) had been systematically measured by Dr. Turan in a prior study. He found that these measurements have similar, albeit slightly lower, sensitivity and specificity compared with the sagittal measurement (personal communication).
Ultrasonographers were instructed on how to correctly identify the closest adrenal gland in the transverse plane. They were then to measure the length (Length) and width (Width) of the total gland as well as the length (length) and width (width) of the fetal zone. Calipers were to be placed directly over the intersection of the gland with the surrounding soft tissues and the intersection of the fetal zone and surrounding gland. Both the length and width of the zone and total gland were measured along the same meridian. This allowed the ratio of the fetal zone to that of the total gland length and width (length/Length and width/Width) to be accurately determined (Fig. 1). By examining the ratio of the fetal zone to the total gland we are able to mitigate differences in measurement that may be the result of fetal size or gestational age.13
Ultrasonographers were instructed to take three measurements of the glands' width and length with the mean values (zone, total, and ratio) used for the analysis. Credentialed ultrasonographers universally completed an online didactic session followed by successful submission of five serial examinations that were blindly reviewed centrally. A total of 29 ultrasonographers were successfully credentialed across the eight clinical sites.
To ensure ongoing quality, every 10th examination was centrally reviewed and, if an inadequate examination occurred, feedback was provided to the ultrasonographer and a progressive system of central review was provided. Ultrasonographers were permitted to spend up to 20 minutes per patient completing the examination; if they were unable to obtain adequate images, they were instructed to designate the study as incomplete. No specifications were placed on the type of ultrasound unit to be used.
The sample size goal for the study was 1,500–2,000 women with measurements based on precision estimates for the area under a receiver operating characteristic curve (AUC). It was assumed that 9% of women would experience a spontaneous preterm birth, and standard errors (SEs) for AUCs of 0.80 and 0.65 were computed. An AUC of 0.65 was considered conservative and a sample size of 1,500 provided an SE of 0.027 and 2,000 provided an SE of 0.023.
Nulliparous women with a singleton fetus enrolled in the parent study undergoing ultrasonography between 22 and 0/7 and 30 6/7 weeks of gestation were recruited into this study. This timeframe was the last prespecified visit in the parent study and was chosen because it was most proximal to the time of preterm birth. All participants had a certain due date corroborated or established by an ultrasonogram before 14 0/7 weeks of gestation.15 Women with uterine anomalies, hydramnios (amniotic fluid index 25.0 cm or greater), history of three or more prior spontaneous losses at less than 20 weeks of gestation, presumptively lethal fetal malformations, or known or suspected aneuploidy (eg, malformations strongly linked with aneuploidy such as congenital diaphragmatic hernia) were excluded. Women who had undergone multifetal reduction, received a cerclage at any point, or received progesterone after 15 weeks of gestation were also excluded. Those women deemed noncompliant with prior study procedures were similarly not enrolled in this portion of the study.
Our primary outcome was spontaneous preterm birth, which was defined as delivery occurring before 37 0/7 weeks of gestation after spontaneous onset of preterm labor or preterm premature rupture or the membranes or fetal membrane prolapse (membranes beyond the cervical os) regardless of subsequent labor augmentation or cesarean delivery. Other prespecified outcomes included spontaneous preterm birth before 34 0/7 weeks of gestation and delivery within 2 weeks of the fetal adrenal measurement.
Analysis consisted of descriptive statistics, including sensitivity and specificity for predicting spontaneous preterm birth over the range of possible cutoffs of the ratio measures and computation of the area under each receiver operating characteristic curve. Wilcoxon rank-sum tests were used to test for differences in location of the data (eg, differences in medians). A P value <.05 (two-sided) was viewed as statistically significant. Intraclass correlation coefficients were computed from the three repeated measurements the ultrasonographers were instructed to obtain. All analyses were conducted using SAS 9.3/9.4.
A total of 2,108 women underwent an attempt at assessment of the fetal adrenal gland, with 1,723 (81.7%) having a successful measurement (Fig. 2). As part of the ongoing central review, 95 studies were excluded and 290 exceeded the 20 minutes allowed to obtain an adequate image. Thus, satisfactory studies were obtained in 81.7% of attempts. Women for whom measurements could not be obtained tended to have higher body mass indexes (calculated as weight (kg)/[height (m)]2) than those with measurements; 33.2% of those without a measurement had body mass indexes of 30.0 or higher compared with 19.9% of those with a measurement (P<.001). Nonetheless, adequate measurements were obtained in the majority of obese women (78.0%). Normative values of measurement by gestational age are available in Table 1.
Pregnancy outcomes were available for 1,697 (98.5%) women with one or more fetal adrenal gland measurements. Spontaneous preterm birth before 37 weeks of gestation was experienced by 82 women (4.8%) and before 34 weeks of gestation by six women (0.4%). There were 1,566 term births. The remaining births were 49 indicated preterm births including one stillbirth, which was censored for the primary analysis.
Baseline characteristics are presented in Table 2 for the 1,723 women with fetal adrenal gland measurements. The median age was 27 years (interquartile range 22–31). The race–ethnicity distribution was 55.7% non-Hispanic white, 12.4% non-Hispanic black, 22.9% Hispanic, 5.4% Asian, and 3.6% other. This was a first pregnancy for 73.9% of the women.
Table 3 provides descriptive statistics for the fetal adrenal gland measurements and a context on the timing with respect to the pregnancy for women with spontaneous preterm birth (n=82) and term birth (n=1,562). Ultrasound assessment of the fetal adrenal gland occurred at a mean of 27.4 weeks of gestation (±1.5 standard deviation). No women delivered within 2 weeks of the time of ultrasound assessment of the fetal adrenal gland. The mean time elapsed between ultrasound measurements of the adrenal gland and preterm delivery was 55.3 (±13.9 standard deviation) days (range 18–93 days).
Medians (interquartile ranges) for length, width, and depth of the total gland, fetal zone, and ratio are displayed in Table 3. None of the measurements differed significantly between women who delivered prematurely and those who delivered at term. Receiver operating characteristic curves and AUCs are shown in Figure 3 for the ratio measures as predictors of spontaneous preterm birth. Panel 1 contrasts spontaneous preterm birth at less than 37 0/7 weeks of gestation with term birth, and, to reflect a pure predictive framework, Panel 2 contrasts spontaneous preterm birth at less than 37 0/7 weeks of gestation with all other births. None of the AUCs were statistically different from 0.50, suggesting that measurements of the ratio of the fetal adrenal zone to the overall gland does not discriminate women who are destined to spontaneously deliver prematurely from those who do not. The AUCs (95% confidence intervals) for spontaneous preterm birth at less than 37 0/7 weeks of gestation were 0.51 (0.45–0.58), 0.50 (0.44–0.56), and 0.52 (0.41–0.63) for width/Width, length/Length, and depth/Depth ratios, respectively, and for spontaneous preterm birth at less than 34 0/7 weeks of gestation were 0.52 (0.25–0.79) and 0.55 (0.31–0.79) for width/Width and length/Length ratios, respectively. There was only one spontaneous preterm birth at less than 34 0/7 weeks of gestation with a depth/Depth ratio measure.
Intraclass correlation coefficients of the ratios for the measurements repeated three times by the ultrasonographer were 0.70, 0.74, and 0.77 for width/Width, length/Length, and depth/Depth, respectively. The median (interquartile range) on the difference between these ratio measurements (absolute value of the maximum difference among the three measurements) were 9.6% (9.4%), 7.9% (8.0%), and 8.6% (8.5%) for width/Width, length/Length, and depth/Depth, respectively.
In contrast to the two prior studies by Turan et al in symptomatic women, the size of the fetal adrenal gland was not predictive of preterm birth in asymptomatic women.12,13 This finding is consistent with an investigation performed by Salari,16 who likewise examined the fetal adrenal gland in a cohort of asymptomatic women. Several explanations may be offered as to the reason for these discrepant findings.
Foremost is the fact that we examined asymptomatic women who averaged 55 days from the time of their ultrasonogram to their preterm birth. It is conceivable that the changes of the adrenal gland occur over a shorter duration and the value of the adrenal gland measurement is only as an acute (within 5–7 days) marker of spontaneous preterm birth rather than an early marker such as cervical length.7
This finding is surprising because maternal estriol levels, which are primarily derived from the fetal adrenal gland, have been noted to rise approximately 4 weeks before the onset of preterm birth.17,18 Moreover, other investigators found that maternal corticotropin-releasing hormone levels (a surrogate marker of the size of the fetal adrenal) begin to rise at 16–20 weeks of gestation in women who go on to have preterm birth.19
Also, it is well known that preterm birth is a common outcome of four antecedent pathways: inflammation, myometrial stretch, activation of the maternal–fetal hypothalamic–pituitary–adrenal axis, and decidual hemorrhage.20 It is conceivable that changes in the fetal zone may not occur if the pathway to preterm parturition is not through activation of the maternal–fetal hypothalamic–pituitary–adrenal axis. Meaningful attribution to these four different pathways has yet to be parsed out. Moreover, the fact that preterm birth has these four antecedent pathways dramatically adds to the complexity of screening in low-risk populations. In addition, each of these pathways may have different biomarkers that are expressed at different gestational epochs, making the discrimination of women who deliver at term from those who deliver preterm very challenging.
Strengths of our study include the very large sample size and multicenter approach. Furthermore, ongoing assessment of ultrasound quality was closely followed throughout the duration of the study. Limitations of our study include the fact that the rate of spontaneous preterm birth was smaller than anticipated and that no woman delivered within 2 weeks of the time that she had the ultrasound measurement. The parent study prospectively recruited women seeking care in the first trimester. This resulted in a cohort of parturients of a higher socioeconomic status and lower risk for preterm birth than that found in the general population.21 Our outcome of interest was spontaneous preterm birth, which excludes women with iatrogenic prematurity. Iatrogenic prematurity typically represents one third of preterm births.22 Spontaneous preterm birth occurred at a rate of 5% in our study population. It is noteworthy that the precision of the AUC measurements that were planned with an assumption of 9% spontaneous preterm births is still adequate using a 5% spontaneous preterm birth rate. The SE for an AUC of 0.65 would be 0.035 (at 5%) instead of 0.027 (at 9%) for n=1,500 and 0.030 instead of 0.023 for n=2,000. Future studies of the fetal adrenal gland may yield different results for high-risk groups and a shorter timeframe between measurement and the preterm birth. Nonetheless, as a single early marker, measurement of the fetal adrenal gland does not appear to predict preterm birth in asymptomatic women.
1. Hamilton BE, Martin JA, Ventura SJ. Births: preliminary data for 2012. Natl Vital Stat Rep 2013;62:1–20.
2. Russell RB, Green NS, Steiner CA, Meikle S, Howse JL, Poschman K, et al.. Cost of hospitalization for preterm and low birth weight infants in the United States. Pediatrics 2007;120:e1–9.
3. Fonseca EB, Celik E, Parra M, Singh M, Nicolaides KH; Fetal Medicine Foundation Second Trimester Screening Group. Progesterone and the risk of preterm birth among women with a short cervix. N Engl J Med 2007;357:462–9.
4. Owen J, Hankins G, Iams JD, Berghella V, Sheffield JS, Perez-Delboy A, et al.. Multicenter randomized trial of cerclage for preterm birth prevention in high-risk women with shortened midtrimester cervical length. Am J Obstet Gynecol 2009;201:375.e1–8.
5. Hassan SS, Romero R, Vidyadhari D, Fusey S, Baxter JK, Khandelwal M, et al.. Vaginal progesterone reduces the rate of preterm birth in women with a sonographic short cervix: a multicenter, randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol 2011;38:18–31.
6. Petrini JR, Callaghan WM, Klebanoff M, Green NS, Lackritz EM, Howse JL, et al.. Estimated effect of 17 alpha-hydroxyprogesterone caproate on preterm birth in the United States. Obstet Gynecol 2005;105:267–72.
7. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, et al.. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network. N Engl J Med 1996;334:567–72.
8. Meis PJ, Klebanoff M, Thom E, Dombrowski MP, Sibai B, Moawad AH, et al.. Prevention of recurrent preterm delivery by 17 alpha-hydroxyprogesterone caproate. N Engl J Med 2003;348:2379–85.
9. Liggins GC, Fairclough RJ, Grieves SA, Kendall JZ, Knox BS. The mechanism of initiation of parturition in the ewe. Recent Prog Horm Res 1973;29:111–59.
10. Anderson AB, Laurence KM, Davies K, Campbell H, Turnbull AC. Fetal adrenal weight and the cause of premature delivery in human pregnancy. J Obstet Gynaecol Br Commonw 1971;78:481–8.
11. Cunningham FG, Leveno KJ, Bloom SL, Spong CY, Dashe JS, Hoffman BL, et al., editors. Williams Obstetrics, 24th Edition. New York: McGraw Hill (2014), 108–110.
12. Turan OM, Turan S, Funai EF, Buhimschi IA, Copel JA, Buhimschi CS. Fetal adrenal gland volume: a novel method to identify women at risk for impending preterm birth. Obstet Gynecol 2007;109:855–62.
13. Turan OM, Turan S, Funai EF, Buhimschi IA, Campbell CH, Bahtiyar OM, et al.. Ultrasound measurement of fetal adrenal gland enlargement: an accurate predictor of preterm birth. Am J Obstet Gynecol 2011;204:311.e1–10.
14. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84.
15. Haas DM, Parker CB, Wing DA, Parry S, Grobman WA, Mercer BM, et al.. A description of the methods of the Nulliparous Pregnancy Outcomes Study: monitoring mothers-to-be (nuMoM2b). Am J Obstet Gynecol 2015;212:539.e1–24.
16. Salari K, Treadwell M, Thompson K. Predictors of preterm birth: a comparison of cervical length, fetal fibronectin and fetal adrenal gland enlargement. Am J Obstet Gynecol 2015;212:S179–180.
17. Ellis MJ, Livesey JH, Inder WJ, Prickett TC, Reid R. Plasma corticotropin-releasing hormone and unconjugated estriol in human pregnancy: gestational patterns and ability to predict preterm delivery. Am J Obstet Gynecol 2002;186:94–9.
18. McGregor JA, Jackson GM, Lachelin GC, Goodwin TM, Artal R, Hastings C, et al.. Salivary estriol as risk assessment for preterm labor: a prospective trial. Am J Obstet Gynecol 1995;173:1337–42.
19. McGrath S, McLean M, Smith D, Bisits A, Giles W, Smith R. Maternal plasma corticotropin-releasing hormone trajectories vary depending on the cause of preterm delivery. Am J Obstet Gynecol 2002;186:257–60.
20. Lockwood CJ, Kuczynski E. Risk stratification and pathological mechanisms in preterm delivery. Paediatr Perinat Epidemiol 2001;15(suppl 2):78–89.
21. Glinianaia SV, Ghosh R, Rankin J, Pearce MS, Parker L, Pless-Mulloli T. No improvement in socioeconomic inequalities in birthweight and preterm birth over four decades: a population-based cohort study. BMC Public Health 2013;13:345.
22. Gyamfi-Bannerman C, Ananth CV. Trends in spontaneous and indicated preterm delivery among singleton gestations in the United States, 2005–2012. Obstet Gynecol 2014;124:1069–74.